NVIDIA Quadro Plex 2200 D2

NVIDIA Quadro Plex 2200 D2 — 4 Гб ×2 (8 Гб) GDDR3, 240 ×2 (480) ядер

Loading...

Performance Rating

The Quadro Plex 2200 D2 was a professional graphics card by NVIDIA, launched on July 25th, 2008. Built on the 55 nm process, and based on the GT200B graphics processor, the card supports DirectX 11.1. Even though it supports DirectX 11, the feature level is only 10_0, which can be problematic with many DirectX 11 & DirectX 12 titles. The GT200B graphics processor is a large chip with a die area of 470 mm² and 1,400 million transistors. Quadro Plex 2200 D2 combines two graphics processors to increase performance. It features 240 shading units, 80 texture mapping units, and 32 ROPs, per GPU. NVIDIA has paired 8 GB GDDR3 memory with the Quadro Plex 2200 D2, which are connected using a 512-bit memory interface per GPU (each GPU manages 4,096 MB). The GPU is operating at a frequency of 648 MHz, memory is running at 800 MHz.

H200 H200
MI325X MI325X
A100 A100

NVIDIA Quadro Plex 2200 D2

NVIDIA Quadro Plex 2200 D2

Memory

Memory Size

4 GB ×2 (8 ГБ)

Memory Type

Memory Bandwidth

102.4 GB/s ×2 (204.8 GB/s)

Memory Bus Width

512 бит ×2 (1024 бит)

ML Performance

FP16 (Half Precision)

No TFLOPS
1.1904 TFLOPS
(Intel UHD Graphics 730)

BF16 (Brain Float)

No TFLOPS
311.84 TFLOPS
(NVIDIA A800 SXM4 80 GB)

TF32 (TensorFloat)

Compute Power

FP32 (Single Precision)

0.6221 TFLOPS

FP64 (Double Precision)

0.0778 TFLOPS
1.2038 TFLOPS
(AMD Radeon RX 7900M)

CUDA Cores

240 ×2 (480)

Architecture & Compatibility

GPU Architecture

Tesla 2.0

SM (Streaming Multiprocessor)

PCIe Version

PCIe 2.0 x16

ML Software Support

CUDA Version

Clocks & Performance

Base Clock

Memory Clock

Power Consumption

TDP/TGP

640 W
unknown
(NVIDIA CMP 70HX)

Recommended PSU

Power Connector

Rendering

Texture Units (TMU)

80 ×2 (160)

Additional

Slots

No
SXM Module
(NVIDIA H200 SXM 141 GB)

Release Date

July 25, 2008

Display Outputs

No outputs
4x mini-DisplayPort 2.0
4x HDMI 2.1
(SPARKLE Arc A310 OmniView)

Recommendations

Эта видеокарта лучше всего подходит для профессиональных задач, таких как рендеринг, 3D-моделирование и визуализация. Она не предназначена для гейминга или обработки больших объемов данных с использованием машинного обучения и AI.

Renting is cheaper than buying